I'm trying to use the arduino to switch a IRL540 MOSFET, which connects a coil to a 50V high current power supply. I've added a 10k pull down resistor across the MOSFETs gate and source, as well as a diode across the coil to prevent reverse current through the diode. All as shown in the attached (the 5V supply and switch is supposed to be a digital output on the arduino).
However, this does not work and I am not sure why. Once I power the gate of the MOSFET it shorts drain and source and remains shorted even when gate voltage is returned to 0, the MOSFET dies.
Any help here greatly appreciated, I feel like I am making a simple stupid mistake.
As drawn it should work assuming your diagram is exactly the way it is assembled and your Arduino Ground and 50 Volt Ground are common (bonded). A gate turn on (current limiting) resistor could be added but I do not see that as preventing the circuit from working. I would start looking for a problem in the wiring and making sure you have the IRL540 pin out correct.
First and foremost whats the resistance of the coil?
You have no protection from dV/dt on the gate.
You have no decoupling on the 50V supply to the coil either.
A 6.8V to 12V zener between gate and source will prevent the gate being pulled too high by
capacitive coupling between drain and gate.
Normally if a MOSFET is controlling high power load you need to drive the gate from low impedance,
and a physical switch is not low impedance when its open... Gate driver chips are the obvious approach
they are made specifically for driving MOSFET gates.
The 50V supply may have big spikes on it without decoupling (which needs to be from the top of
the coil direct to ground at the MOSFET source).
First and foremost whats the resistance of the coil?
2 Ohms
You have no decoupling on the 50V supply to the coil either.
The 50V supply is actually a 18650 bank intended only to power the coil for about 20ms at a time, so I assumed any small fluctuations in voltage wouldn't be very important.
A 6.8V to 12V zener between gate and source will prevent the gate being pulled too high by
capacitive coupling between drain and gate.
I'm still a begginer so I'm not sure what this means. How would the gate be pulled too high? Too high, as in above the maximum gate-source voltage (16V)?
Normally if a MOSFET is controlling high power load you need to drive the gate from low impedance,
and a physical switch is not low impedance when its open... Gate driver chips are the obvious approach
they are made specifically for driving MOSFET gates.
I wasn't sure how to represent an Arduino digital output in CircuitLab so I used a button instead. I'm powering the gate using an Arduino in reality.
A gate turn on (current limiting) resistor could be added
I did actually include one, I just forgot to add it in the schematic. Whoops.
Show us a good image of your 'actual' wiring.
Give links to components.
The components are as noted in the schematic, except the coil. My wiring would be pretty indecipherable from an image, but I have checked again and again and my wiring seems to be correct.
Something I should probably have mentioned initially is that there are multiple of these modules connected to the same power supply.
So 50V/2 ohms = 25A. That MOSFET's on-resistance is 53 miiliohms, so 25A means 33W in the MOSFET
and 1.25kW in the coil, assuming the 50V holds up. Are you intending for such massive power draw?
I'm still a begginer so I'm not sure what this means. How would the gate be pulled too high? Too high, as in above the maximum gate-source voltage (16V)?
With rapidly switching high voltages there is coupling via device capacitances between drain and gate that
can destroy the gate. You have to hold the gate voltage solidly (10k resistor is not solid). A zener diode
will prevent the gate voltage going too high or too low.
A typical MOSFET turning on in this circuit might have dV/dt on the drain at 100 to 500 volts/microsecond,
easily enough to push many milliamps into the gate via capacitance. Switch-off can be even worse with
an inductive load.
This all happens on a timescale of nanoseconds, you need an oscilloscope to see whether this
is really a problem - just adding a zener is a simple and easy fix.
I wasn't sure how to represent an Arduino digital output in CircuitLab so I used a button instead. I'm powering the gate using an Arduino in reality.
Well that would have been really useful to know...
The 50V supply is actually a 18650 bank intended only to power the coil for about 20ms at a time, so I assumed any small fluctuations in voltage wouldn't be very important.
I'm still a begginer so I'm not sure what this means. How would the gate be pulled too high? Too high, as in above the maximum gate-source voltage (16V)?
I wasn't sure how to represent an Arduino digital output in CircuitLab so I used a button instead. I'm powering the gate using an Arduino in reality.
I did actually include one, I just forgot to add it in the schematic. Whoops.
The components are as noted in the schematic, except the coil. My wiring would be pretty indecipherable from an image, but I have checked again and again and my wiring seems to be correct.
Something I should probably have mentioned initially is that there are multiple of these modules connected to the same power supply.
In your second drawing, the diodes are backwards and the coil is never energised.
What is the coil anyway? A solenoid, relay, motor?
In your first post you said: "as well as a diode across the coil to prevent reverse current through the diode. "
You have this exactly backwards- the diode is there to short the reverse current through the diode.
